Michael Burst – Mediengruppe RTL

Mediengruppe RTL is the leading provider of video content in Germany with 14 station brands and the TVNOW streaming service. In his position as Vice President, Michael Burst is responsible for Content & Data Analytics at Data Alliance, the Data Competence Center of Mediengruppe RTL as well as Gruner and Jahr. He leads four teams that, among other things, take care of the quantitative data provision, run analyses and provide insights by creating reports, dashboards and models. His teams’ main task is to provide decision-makers, programme creators and editors with data-driven, actionable recommendations. His department is responsible for managing the complete data path required for this.

User preference & attention as KPIs for growth

Within Mediengruppe RTL, the most important control parameters include key figures such as the number of viewers, users and session duration per user. For TVNOW, RTL Group’s streaming service, Business KPIs also include subscriber numbers, growth rates, NPS (Net Promoter Score) and churn rates. Michael Burst explains: “There are different KPIs and it depends a lot on which glasses we are wearing. We look after the publisher side as well as the advertising marketing side. From the publisher’s point of view, for example, the number of people who use our offers stands above all. This means that the classic KPIs such as the number of users & viewers, but also subscribers, typically count. The session duration is pivotal for advertising marketing. Ultimately, this is what we all strive for in the media sector, we all fight for the user’s preference and attention. Every second and every minute that an end-user spends on our pages helps us to understand him or her better.”

“We all fight for the user’s preference and attention. Every second and every minute that an end-user spends on our pages helps us to understand him or her better.”

Becoming the tech & data powerhouse

When it comes to data-driven decision-making, the Mediengruppe RTL is at the forefront of many areas, states Michael Burst. And that is mainly due to the company’s clearly defined strategy and vision for the future: “Two years ago, management decided that they wanted to develop from a pure content company into a content, tech & data company. This led to a change within the company, a change in the value system. And the whole thing works because there was a clear decision and a strong commitment to do this at the top management level. Next to that, there is a clear strategy and a vision for the future which provides everyone with the necessary orientation. Since then, we have invested a lot in technology, but also very heavy in brainpower, including data scientists, data engineers and data analysts.”

Michael adds: “A few years ago, the industry didn’t know exactly what data scientists were doing. The challenge today is how to recruit these experts. In the long term, we want to be a company that produces several AI or automated data solutions every year. At the end of the day, that will bring us significant competitive advantages.”

“A few years ago, the industry didn’t know exactly what data scientists were doing. The challenge today is how to recruit these experts.”

Big data becomes smart data

“A good example of where the journey is headed is our latest product Brand Aid”, Michael Burst shares. He explains: “This is an AI-supported advertising impact tool. With that tool, we were able to create an automatic advertising impact analysis. It intelligently uses the possibilities of big data and machine learning. The Brand Aid dashboard, which is available to the client, is also their control tool. In the future, advertisers will also have the opportunity to use a forecast model. This will enable them to see what exactly will happen to the brand KPIs if they for example invest more in print, TV, etc. For me, this new product is a great example, because it brings together topics such as big data, machine learning, AI and advertising impact requirements. Three, four years ago, this would not have been possible. This is our first product of this kind, but we are currently working on a few more.”

Data-driven & gut feeling

Should decisions entirely be made by machines? Or do you still need some level of gut feel? For Michael Burst it is a combination of both: “I use the gut feeling analogy very often in my day-to-day business. My goal is to lower the share of gut feel when deciding with the help of our data and our analyses so that this decision is data-based. The data-based share of a decision should gain in importance. However, this should not mean that the gut feeling disappears completely. Because that’s what defines us, humans, with our many years of experience. We want to use systems that are automated to the extent that we can free people from routine work. Instead, we enable them to invest their time in things that will help us move forward even more efficiently.”

Machines cannot completely take over decisions, states Michael Burst, at least considering what is possible nowadays. But, in his opinion, machines can be extremely helpful in decision-making: “It is an interaction. The advantage of the machine lies in the fast and efficient processing of many different data sources. But what the machine delivers must be interpreted by a subject matter expert, someone with experience. Because not every result that is produced by a machine or an algorithm has to be implemented, you have to be able to explain it and understand how it comes about.”

Data products as a competitive advantage

Together with his teams, Michael Burst works developing new competitive advantages daily. “In the end, the data-driven solutions that we develop are thoroughly challenged. We usually develop data products that need to create a competitive advantage and ultimately generate more sales. For instance, when fewer people cancel our services because we are using machine learning to improve our recommendation system. If we manage to hold the viewer for even a few minutes longer and increase the session duration, we will receive additional new data points. These data points will ideally help us to optimise the system and the algorithm and thus increase retention and loyalty.”

What should you watch out for?

Due to this dynamic and rapid development, companies must be careful not to fall into a naïve faith in technology or an action mania, concludes Michael Burst. A clear focus and goal orientation are essential. After all, a heap of data alone will not lead to anything at first: “You can compare it to the stock market. You need to think carefully about when to act or change your strategy. Especially in marketing, where you can’t change decisions every day. It takes time for a marketing message to break through and for that to be reflected in the numbers. Therefore, more than 5-7 clearly defined KPIs make no sense. And if you have too much data, you run the risk of not being able to cope with the complexity.”

Michael Burst draws inspiration from companies such as Google, Amazon, Netflix, Facebook and Apple, who recognised how important it is to have control over your own data at a very early stage, and started to set up specialist departments early.